Print This Post Print This Post

The results of the 2008 Research Assessment Exercise were released today. Quality profiles for all units of assessment can be found on the RAE website at:
I’m not quite sure yet how to interpret the results. If one looks just at the percentage of 4*, which is categorized as research ‘that is world-leading in terms of originality, significance and rigour,’ then it would appear that UCL and St. Andrews would come out on top. But it seems that theoretically, a department could just submit one person and get 100% of 4* and come out on top. So it seems that the number of people submitted should also matter. But if that is the case, Oxford, which has the largest philosophy faculty in the UK, and which also submitted the largest percentage of its staff, would seem to come out on top using this metric (ie, the figure for FTE Category A staff submitted times the percentage of research activity at quality level 4*). What’s the right metric to use?


  1. 1. Posted by Nick Shackel | December 19, 2008 1:28 am

    For the sake of argument, let’s make the big assumptions that the figures are roughly accurate and that the goodness of the products is measured (not just ordered) by the categories (e.g. so 4* products are 4 times as good as 1*). Then the mean multiplied by the number of staff gives you a figure directly proportional to the gross value(and I think that the figure so obtained is what is meant by the ‘research power’ terminology of the RAE) Does such an order based on that figure tell us something of what we want to know? Perhaps– it does tell us which universities produced most research value (granted the heavy assumptions). But it ignores opportunity cost –what else might we have done with these resources–so on its own it can’t tell us whether producing this research value was worth it.

    An interesting figure would be to divide the gross value through by the actual costs, or perhaps just employers’ costs of the staff, to find out value for money, and produce a ranking on that basis. In this case we might want to include the costs of all staff who have a research element in their contract, not just all those who were submitted, and salaries scaled according to the research element.

  2. 2. Posted by S. Matthew Liao | December 19, 2008 7:46 pm

    Another reason to take the RAE results with a grain of salt might be the following: The process asks an individual to submit his/her four best pieces of work. Suppose though that an individual has more than four pieces of work that would garner a 4*. For example, let us supppose that an individual has 10 pieces of work that would get 4*’s if she were able to submit them all. On the present RAE system, she would be rated equally as another individual who, let us suppose, had only 4 pieces of work that would merit 4*’s.

Post a comment

Name: (required)

Email Address: (required) (will not be published)